Sex differences in gene expression related to antipsychotic induced weight gain

Abstract

Antipsychotics are crucial for the treatment of schizophrenia and contribute to weight gain in psychosis, particularly during early phases. Antipsychotic Induced Weight Gain (AIWG) might contribute to reduce the quality of life, drug compliance and to increase mortality. To characterize sex differences of gene expression related to AIWG, we sequenced total mRNA from blood samples of schizophrenia patients, before and after 3 months of antipsychotic-treatment. We analyzed schizophrenia patients according to their sex (38 males and 39 females) and their BMI increase after medication, characterizing the differential gene expression before and after medication. Individuals in each group were categorized in patients who gain weight and those whose do not gain weight. The “weight gain” groups included patients with an increase of body mass index (BMI) > 1.0 points (27 males and 23 females with a median BMI increase of 2.68 and 2.32 respectively). The “no weight gain” groups included patients with a change of BMI between < 1.0 and > -1.0 points (11 males and 16 females with a median BMI increase of 0.21 and 0.16 respectively). The males had 331 genes with significant differential expression in the weight gain group and 24 genes in the no weight gain group. The females had 119 genes with significant differential expression in the weight gain group and 75 genes in the no weight gain group. Both weight gain groups were significantly enriched with “obesity” genes (Fisher; p = 1.1E-09 and p = 0.0001 respectively), according to the Gene Reference into Function (GeneRIF) database.In conclusion, we characterized genes with differential expression associated to AIWG that are specific to males, to females and common to both sexes. These genes are good candidates to depict the biological processes involved in AIWG and provide additional evidence of the genetic links between weight gain and the immune system.

About the Authors:

Jesus Sainz

Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Supervision, Validation, Writing – original draft, Writing – review & editing

* E-mail: [email protected] (JS); [email protected] (BC-F)

Affiliation: Spanish National Research Council (CSIC), Institute of Biomedicine and Biotechnology of Cantabria (IBBTEC), Santander, Spain

ORCID logo http://orcid.org/0000-0003-4018-7175

Carlos Prieto

Roles Data curation, Formal analysis, Funding acquisition, Methodology, Software, Writing – review & editing

Affiliation: Bioinformatics Service, Nucleus, University of Salamanca (USAL), Salamanca, Spain

Benedicto Crespo-Facorro

Roles Funding acquisition, Investigation, Resources, Supervision, Validation, Writing – review & editing

* E-mail: [email protected] (JS); [email protected] (BC-F)

Affiliations University Hospital Marqués de Valdecilla, IDIVAL, Department of Psychiatry, School of Medicine, University of Cantabria, Santander, Spain, CIBERSAM, Centro Investigación Biomédica en Red Salud Mental, Santander, Spain, University Hospital Virgen del Rocio, University of Sevilla, Seville, Spain

Introduction

Antipsychotic medications are the keystone for the treatment of schizophrenia and other psychotic disorders (including schizoaffective disorder, delusional disorder and bipolar affective disorder)[1], although they are also extensively used in several other disorders such as personality disorders, dementia and autism[2–4]. The clinical differences between antipsychotic drugs are mainly in the areas of safety and tolerability. The second-generation antipsychotics (SGAs) as a group are recommended, due to a lesser association with extrapyramidal symptoms and hyperprolactinemia, by most guidelines as first-line therapy for schizophrenia[5]. Although several other factors (i.e.; genetic susceptibility, sedentary lifestyle and unhealthy food habits) may account for weight gaining in patients with psychosis, antipsychotic treatment (especially SGAs) is considered a significant contributor to weight gain, in particular during early phases of psychosis[6–8]. In consequence, antipsychotic-induced weight gain (AIWG) is a major management problem for clinicians, given that obesity and overweight have a strong negative impact in health[9]. Weight gain is a risk factor for diabetes type II and can lead to metabolic syndrome, a severe condition with serious implications for survival[10]. Thus, the consequences of a long-term treatment with antipsychotics comprise not only overweight and obesity, but also metabolic disturbances and cardiovascular disease[11, 12], although cardiovascular mortality in schizophrenia is attributable to other factors than antipsychotic treatment when used in adequate dosages[13]. The reduced life expectancy for schizophrenic patients is attributable in a great extent to a higher rate of cardiovascular disease[14–17]. There are great differences in obesity according to sex. Overall, more women are obese than men, yet, in developed countries, more men are overweight than women[18]. The reasons of gender differences in obesity and the molecular mechanisms behind AIWG are not well known. Given that that AIWG has a strong genetic component[19], we analyzed the transcriptome expression changes related to weight gain caused by antipsychotics. To obtain information about the genetics behind the sex differences, we analyzed independently males and females comparing individuals with weight gain and without it. We sequenced the blood transcriptome of male and female cohorts of first episode schizophrenia patients before (antipsychotic naïve) and after three months of treatment with antipsychotics (Table 1). The “weight gain” cohorts included 27 patients in the male group and 23 patients in the female group that gained more than 1.0 points of BMI after the treatment (Fig 1). The “no weight gain” groups included 11 males and 16 females with a change on BMI after medication lower than 1.0 points and higher than -1.0 points. The transcriptomes of the four different groups were analyzed independently to define genes with significant differential expression before and after medication in the male and female weight and no weight gain groups using the program Deseq[20].

[Figure omitted. See PDF.]

Fig 1. Whisker plot of body mass index (BMI) data in the male and female weight gain groups.

https://doi.org/10.1371/journal.pone.0215477.g001

[Figure omitted. See PDF.]

Table 1. Anthropometric and metabolic values in the groups studied.

https://doi.org/10.1371/journal.pone.0215477.t001

Materials and methods

Study setting and subjects

The cohort analyzed in this study was obtained at the University Hospital Marques de Valdecilla (Cantabria, Spain). Conforming to international standards for research ethics, this study was approved by the Cantabria Ethics Institutional Review Board (IRB). Patients meeting inclusion criteria and their families provided written informed consent to be included in the study. The biological samples of patients were provided by the Valdecilla biobank (Santander). All methods were performed in accordance with the relevant guidelines and regulations. Before giving informed consent, all patients were evaluated by a clinician to assess their competency (according to their capacity of understanding, reasoning, and expression of a choice). After informed consent was signed, patients were included in a prospective, randomized, flexible-dose, open-label study.

All patients included in the study met the following criteria: 1) 15–60 years old; 2) living in the catchment area (Cantabria); 3) experiencing a first episode of psychosis; 4) having received no prior treatment with antipsychotic medication; 5) DSM-IV criteria for schizophrenia, schizophreniform disorder, schizoaffective disorder, or brief psychotic disorder. Patients were excluded from the study for any of the following reasons: 1) meeting DSM-IV criteria for drug dependence, 2) meeting DSM-IV criteria for mental retardation, 3) having a history of neurological disease or head injury. The diagnoses were confirmed, using the Structured Clinical Interview for DSM-IV (SCID–I), by an experienced psychiatrist after 6 months of the baseline visit. Our operational definition for a “first episode of psychosis” includes individuals with a non-affective psychosis who have not previously received antipsychotic treatment regardless the duration of psychosis.Only individuals who gave written consent to their participation in the program, fulfilled inclusion criteria at 6 months, and had mRNA samples available at baseline and at 3 months were finally included in the present study (S1 Fig.). No statistically significant differences were found in clinical and sociodemographic characteristics at baseline between those patients included (N = 77) and no included (N = 98) in final analyses (all ps > 0.05), except for CGI and BPRS (S1 Table).

Anthropometric and metabolic values (age, body mass index, triglycerides and cholesterol) were collected from patients before and after medication in the groups studied. (Table 1).

Sample and study design

We selected two groups of patients to sequence their mRNA from blood samples before and after 3 months of treatment with antipsychotics. We defined the “weight gain” groups by an increase of BMI > 1.0 points; and the “no weight gain” group by a change of BMI between < 1.0 and > -1.0 points. The “weight gain” cohorts included 27 patients in the male group and 23 patients in the female group. The “no weight gain” groups included 11 males and 16 females.

Laboratory assessments

Blood samples were assessed for biochemical and hematological parameters. Blood samples were obtained from fasting subjects from 8:00 to 10:00 a.m. by the same staff and in the same setting. A detailed description of methodology followed to assess biochemical variables is available under request. None of the patients had a chronic inflammation or infection, or were taking medication that could influence the results of blood tests.

RNA extraction

Total RNA was extracted from blood using the Tempus Blood RNA Tube and Tempus Spin RNA Isolation Kit (Applied Biosystems, Foster City, CA, USA) following the manufacturer protocols. To define expression profiles, a key factor is that the RNA is intact. To select only high-quality RNA, the RNA Integrity Number (RIN) was characterized with a Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) and samples with a RIN of at least 7.4 were selected. We selected this cut-off because it ensures a minimum of quality regarding mRNA integrity. The majority of the selected samples have RINs that range from 8 to 10 with an average of 9.1.

RNA Next generation sequencing

Total RNA was extracted from peripheral blood of each individual. The mRNA obtained from blood was sequenced at the Centro Nacional de Análisis Genómico (CNAG) using Illumina HiSeq instruments (San Diego, CA, USA). The mRNA was isolated from the total RNA and was fragmented once transformed into cDNA. Fragments of 300 bp on average were selected to construct the cDNA libraries for sequencing by the “Centro Nacional de Análisis Genómico” (CNAG). These libraries did not included ncRNA, only mRNA. Pair-end sequences of 70 nucleotides for each end were produced.

Alignment of reads to the human genome reference

Alignment of the reads was performed in an SLURM HPC server running Tophat 2.0.6 with default options[21]. Tophat aligns RNA-Seq reads to genomes using the Bowtie 2.0.2 alignment program[22], and then analyzes the mapping results to identify splice junctions between exons.

Differential expression and statistical analyses

Bedtools 2.17.0 (multicov option)[23] was used to count the amount of reads mapped to each gene. The Reference Sequence (RefSeq) gene coordinates were defined using the RefFlat file from the UCSC Genome Bioinformatics Site (as February 28th, 2014). Deseq 1.4 package[20], setting up fit-only as fitting method, was used to test for differential expression using gene-count data. P value was adjusted for multiple testing with the Benjamini-Hochberg procedure, which controls false discovery rate.

Two sided Fisher tests were carried out to identify functional enrichment of biological annotations. Linear regression analyses of expression changes versus BMI changes were performed without obtaining relevant results.

Results

Differential gene expression in the male “weight gain” group before and after treatment with antipsychotics

We found 331 genes with significant differential expression before and after medication (Padj value <0.01) (S2 Table).

These genes were significantly enriched for the strings “obesity” or “BMI” according to the scientific literature in the GeneRIF database that provides functional and morbid annotation of genes[24].

We found 42 “obesity” related differential expression genes or 14.4% of the annotated genes while we expect 4.9% (Fisher; p = 1,098E-09). When we analyzed for the string “BMI”, we found 15 genes or 5.1% of the annotated genes while we expect 2.8% (Fisher; p = 0.03).

These differential expression genes were also significantly enriched for “cholesterol” related genes, 43 or 14.8% of the annotated genes while we expect 5.2% (Fisher; 1,776E-09); and for “triglyceride” related genes, 15 or 5.1% of the annotated genes while we expect 1.9% (Fisher; p = 0.0005). Additionally, there was a strong enrichment for diabetes genes: 79 or 27.1% while we expect 10.6% (Fisher; p = 8.28E-15).

Differential gene expression in the female “weight gain” group before and after treatment with antipsychotics

We found 119 genes with significant differential expression before and after medication (Padj value <0.01) (S3 Table).

These genes were enriched for the string “obesity” and weakly enriched for “BMI” according to the scientific literature in the GeneRIF database[24].

We found 16 “obesity” differential expression genes or 14.7% of the annotated genes while it was expected 4.9% (Fisher; p = 0.0001). When we analyzed for the string “BMI”, we found 7 genes or 6.4% of the annotated genes while it was expected 2.8% (Fisher; p = 0.03). There was also a strong enrichment for diabetes genes: 30 or 27.5% while we expect 10.6% (Fisher; p = 8.09E-07).

The differential expression genes were significantly enriched for “cholesterol” related genes: 16 or 14.7% of the annotated genes while we expect 5.2% (Fisher; p = 0.0002) and for “triglyceride” related genes: 7 or 6.4% of the annotated genes while we expect 1.9% (Fisher; p = 0.005). There was also a strong enrichment for “diabetes” genes: 30 or 27.5% while we expect 10.6% (Fisher; p = 8.09E-07).

Genes with differential expression specific to the male and female “weight gain” groups only

By subtracting the differential expression genes of the male and female no weight gain groups, and the female weight gain group (S3, S4 and S5 Tables respectively), we found specifically 250 genes with differential expression in the male weight gain group.

By subtracting the differential expression genes of the male and female no weight gain groups, and the male weight gain group (S2, S4 and S5 Tables respectively), we found 37 genes specific to the female weight gain group.

Common genes to male and female having differential expression only in the “weight gain” groups

We found 47 genes with differential expression common to the male and female weight gain groups (S2 and S3 Tables respectively) and not present in the male nor in the female no weight gain groups (S4 and S5 Tables respectively)

Pathway analyses

Analyses of several pathway databases (Reactome[25], KEGG[26], WikiPathways[27] and Pathway Interaction Database[28]) revealed that pathways related to the immune system were highly enriched with differential expression genes common to the male and female weight gain groups and not present in the no weight gain groups. The three most significantly gene enriched pathways were all immune system related. Neutrophil Degranulation with 17 genes (Observed 42%, expected 4%; Fisher p = 1.06E-13). Innate Immune System with 19 genes (Observed 47%, expected 11%; Fisher p = 5.6E-09). Immune System with 22 genes (Observed 55%, expected 18%; Fisher p = 1.0E-07).

Using the same type of analyses, we found very similar results for the genes specific to the male and the female weight gain groups respectively.

For the genes specific to the male weight gain group, the three most significantly gene enriched pathways were: Immune System with 71 genes (Observed 40%, expected 18%; Fisher p = 3.3E-12). Innate Immune System with 50 genes (Observed 28%, expected 11%; Fisher p = 1.7E-10). Neutrophil Degranulation with 27 genes (Observed 15%, expected 4%; Fisher p = 7.06E-09).

For the genes specific to the female weight gain group, the three most significantly gene enriched pathways were: Interferon alpha/beta signaling with 5 genes (Observed 18%, expected 1%; Fisher p = 7.47E-07). Interferon Signaling with 6 genes (Observed 21%, expected 2%; Fisher p = 6.4E-06). Immune System with 13 genes (Observed 46%, expected 18%; Fisher p = 0.0004).

Weighted gene co-expression network analysis (WGCNA) and linear regression analyses

In the linear regression analyses of expression changes versus BMI changes, we did not found any significant gene (Padj<0.001) in males and only seven genes in females (S6 Table). We also performed a WGCNA (Weighted Gene Correlation Network Analysis) and resulting modules were correlated with BMI changes. Some of these modules were significantly correlated with BMI changes but we did not find a functional enrichment of BMI related functions (S7 and S8 Tables).

Discussion

It is well known that antipsychotics induce weight gain[29] and that genetic factors play a major role in this weight gain.[7]. Our analysis provides genetic expression data that could clarify the genetic component of the weight gain caused by antipsychotics. It is also well known that adolescent patients gain more weight than older patients[15]. In the cohort analyzed here, the median of age of the male and female weight gain groups had a median of age 10.9 and 14.2 years younger, respectively, than the correspondent no weight gain groups (Table 1), consistently with the literature. One hypothesis is that weight gain is more marked and frequent in the younger patients due to less prior antipsychotic exposure compared to the older patients[30]. However, this hypothesis could not explain the differences observed in our cohort given that all are naïve patients. Other explanation is that the younger patients are simply more sensitive to the weight gain caused by antipsychotics[31], though this is not supported by any molecular mechanism underlying it. Here we provide gene functional data that could explain partially the molecular basis of the differences in AIWG between young and adult.

It also has been reported sex differences in the schizophrenia age of onset, being the males age of onset earlier than the females[32]. We observed also in our cohort that males are younger than females (2.0 and 5.3 years younger in the weight and no weight gain groups respectively; Table 1). These sex differences in age could influence the expression profiles we characterized. However, the genes that are specific of the weight gain groups, even if their expression was age-related, they still could explain the molecular mechanisms of obesity given that these changes are associated to antipsychotics intake and to weight gain.

In the male weight gain group, we found 331 genes with differential expression after medication with antipsychotics. Out of the 331 genes, there are 310 genes with expression not altered by medication in the male no weight gain group. After similar analysis for the female groups, we find 94 genes out of the 119 genes with differential expression in the female weight gain group but not in the female no weight gain group.

Among the genes appearing only in the male and female weight gain groups, 47 are common to male and female, representing 50% of the female 94 female weight gain specific genes.

In males, there are 240 genes having altered expression specific to the weight gain group. These 240 male specific genes have no differential expression in any of the other analyses performed (male no weight gain, female weight gain and female no weight gain groups).

In females, we find 37 genes with altered expression specific to the female weight gain group. The expression of these genes was not altered after medication in any of the other three analyzed groups.

Out of the 240 genes with differential expression only in the male weight gain group, 40 genes are related to obesity traits and metabolic syndrome according to the GeneRif literature. The top ten with the most significant differential expression are: H19[33], ADM[34], CES1[35], S100A8[36], ACSL1[37], C3AR1[38], SDC3[39], KLF4[40], IFI30[41] and FABP1[42]. Among the remaining of the 40 genes, the cholinergic receptor muscarinic 3, or CHRM3, has genetics variants that have been associated with BMI after antipsychotic medication[43]. It is interesting to report that two of the remaining genes that do not appear associated to obesity traits in GeneRif are indeed involved in leptin pathways: BCL2L1, BCL2 like 1, part of the Leptin signaling pathway (WikiPathways); and NRG1, neuregulin 1, part of the Signaling by Leptin pathway (Reactome) and also associated to schizophrenia[44].

In females, we find four genes related to obesity traits and metabolic syndrome out of the 37 genes with differential expression specific to the female weight gain group. These genes are CEACAM1[45], SDC1[46], PLA2G2A[47] and HP[48].

Out of the 47 common genes to male and female specific to the weight gain groups, 11 relate to obesity traits and metabolic syndrome according to previous studies: LTF[49], OLFM4[50], LCN2[51], OLR1[52], MMP8[53], PDK4[54], RNASE3[55], APOA4[56], CHIT1[57], GPER1[58] and CPT1A[59]. The gene OLAH, oleoyl-ACP hydrolase, has not been reported as obesity-trait related in GeneRif but is included in several Reactome lipid-related pathways: Metabolism of lipids and lipoproteins, Fatty acid, triacylglycerol, and ketone body metabolism, Triglyceride Biosynthesis and Fatty Acyl-CoA Biosynthesis.

When we search for “obesity” in the GeneRIF database, we find a significant enrichment of genes with altered expression after antipsychotic medication in the male and female weight gain groups (Fisher; p = 1.1E-09 and p = 0.0001 respectively). Regarding the remaining genes not published as related to obesity traits, is likely to include genes with a relation still unknown to obesity traits. The pathways that some of these genes are included point in that direction.

Pathway analyses of the genes with altered expression specific to the weight gain groups (common to male and female and specific to male or to female) indicate a consistent enrichment of immune system pathways. These observations support a relationship of weight gain and the immune system, as it has been previously suggested[60–62].

We observe a higher enrichment of obesity genes with expression altered by antipsychotics in the males than in the females. A possible explanation for this observation could be that the male groups are younger than the female groups. This is consistent with the studies that report that young people are more sensitive to AIWG. However, we cannot exclude other factors such as BMI before medication; males have a median of 1.9 and 6.4 points of higher BMI than the females in the weight gain and no weight gain groups respectively (Fig 1 and Table 1). In conclusion, our study provides expression data of a list of genes that we believe are good candidates to be involved in biological processes associated with AIWG in the patients medicated for psychosis, depression and other diseases. Logically, these genes could be also involved in the molecular mechanisms that lead to obesity traits in the general population. Also provides additional evidence of the genetic links between weight gain and the immune system. Finally, we expect that this study could facilitate creating predictive tools of weight gain and in turns tailoring specific interventions according to individual risks. The data obtained is a first step to generate a tool for clinicians to predict personalized AIWG for each antipsychotic available. Consequently, more studies would be needed to analyze the AIWG for each antipsychotic and to add data from cohorts from different ethnicities.

Supporting information

[Figure omitted. See PDF.]

S1 Fig. Flow chart of patients.

https://doi.org/10.1371/journal.pone.0215477.s001

(PDF)

S1 Table. Comparison of clinical and sociodemographic characteristic between included (N = 77) and no included patients (N = 98).

https://doi.org/10.1371/journal.pone.0215477.s002

(XLSX)

S2 Table. Differential expression in the male weight gain group before and after 3 months of antipsychotic medication.

https://doi.org/10.1371/journal.pone.0215477.s003

(XLSX)

S3 Table. Differential expression in the female weight gain group before and after 3 months of antipsychotic medication.

https://doi.org/10.1371/journal.pone.0215477.s004

(XLSX)

S4 Table. Differential expression in the male no weight gain group before and after 3 months of antipsychotic medication.

https://doi.org/10.1371/journal.pone.0215477.s005

(XLSX)

S5 Table. Differential expression in the female no weight gain group before and after 3 months of antipsychotic medication.

https://doi.org/10.1371/journal.pone.0215477.s006

(XLSX)

S6 Table. Linear regression analyses of expression changes versus BMI changes in females.

https://doi.org/10.1371/journal.pone.0215477.s007

(XLSX)

S7 Table. WGCNA results with male samples.

https://doi.org/10.1371/journal.pone.0215477.s008

(XLSX)

S8 Table. WGCNA results with female samples.

https://doi.org/10.1371/journal.pone.0215477.s009

(XLSX)

Acknowledgments

Bioinformatics work was partially performed using the Altamira supercomputer (Spanish Supercomputing Network). We thank the Valdecilla Biobank for blood RNA sampling handling and storage and the CNAG for mRNA sequencing. We also wish to thank the participants and their families for enrolling in this study.

Citation: Sainz J, Prieto C, Crespo-Facorro B (2019) Sex differences in gene expression related to antipsychotic induced weight gain. PLoS ONE 14(4): e0215477. https://doi.org/10.1371/journal.pone.0215477

References

1. Insel TR. Rethinking schizophrenia. Nature. 2010;468(7321):187–93. Epub 2010/11/12. pmid:21068826.

2. Roberts RJ, Lohano KK, El-Mallakh RS. Antipsychotics as antidepressants. Asia-Pacific psychiatry: official journal of the Pacific Rim College of Psychiatrists. 2016;8(3):179–88. Epub 2015/05/13. pmid:25963405.

3. Alexander GC, Gallagher SA, Mascola A, Moloney RM, Stafford RS. Increasing off-label use of antipsychotic medications in the United States, 1995–2008. Pharmacoepidemiology and drug safety. 2011;20(2):177–84. Epub 2011/01/22. pmid:21254289; PubMed Central PMCID: PMC3069498.

4. Carton L, Cottencin O, Lapeyre-Mestre M, Geoffroy PA, Favre J, Simon N, et al. Off-Label Prescribing of Antipsychotics in Adults, Children and Elderly Individuals: A Systematic Review of Recent Prescription Trends. Current pharmaceutical design. 2015;21(23):3280–97. Epub 2015/06/20. pmid:26088115.

5. Barnes TR. Evidence-based guidelines for the pharmacological treatment of schizophrenia: recommendations from the British Association for Psychopharmacology. J Psychopharmacol. 2011;25(5):567–620. Epub 2011/02/05. pmid:21292923.

6. Correll CU, Lencz T, Malhotra AK. Antipsychotic drugs and obesity. Trends in molecular medicine. 2011;17(2):97–107. Epub 2010/12/28. pmid:21185230; PubMed Central PMCID: PMC3053585.

7. Shams TA, Muller DJ. Antipsychotic induced weight gain: genetics, epigenetics, and biomarkers reviewed. Current psychiatry reports. 2014;16(10):473. Epub 2014/08/21. pmid:25138234.

8. Perez-Iglesias R, Martinez-Garcia O, Pardo-Garcia G, Amado JA, Garcia-Unzueta MT, Tabares-Seisdedos R, et al. Course of weight gain and metabolic abnormalities in first treated episode of psychosis: the first year is a critical period for development of cardiovascular risk factors. Int J Neuropsychopharmacol. 2014;17(1):41–51. Epub 2013/10/10. pmid:24103107.

9. Swinburn BA, Sacks G, Hall KD, McPherson K, Finegood DT, Moodie ML, et al. The global obesity pandemic: shaped by global drivers and local environments. Lancet. 2011;378(9793):804–14. Epub 2011/08/30. pmid:21872749.

10. De Hert M, Detraux J, van Winkel R, Yu W, Correll CU. Metabolic and cardiovascular adverse effects associated with antipsychotic drugs. Nature reviews Endocrinology. 2011;8(2):114–26. Epub 2011/10/20. pmid:22009159.

11. Baker JL, Olsen LW, Sorensen TI. Childhood body-mass index and the risk of coronary heart disease in adulthood. The New England journal of medicine. 2007;357(23):2329–37. Epub 2007/12/07. pmid:18057335; PubMed Central PMCID: PMC3062903.

12. Gebhardt S, Haberhausen M, Heinzel-Gutenbrunner M, Gebhardt N, Remschmidt H, Krieg JC, et al. Antipsychotic-induced body weight gain: predictors and a systematic categorization of the long-term weight course. Journal of psychiatric research. 2009;43(6):620–6. Epub 2008/12/27. pmid:19110264.

13. Torniainen M, Mittendorfer-Rutz E, Tanskanen A, Bjorkenstam C, Suvisaari J, Alexanderson K, et al. Antipsychotic treatment and mortality in schizophrenia. Schizophrenia bulletin. 2015;41(3):656–63. Epub 2014/11/26. pmid:25422511; PubMed Central PMCID: PMC4393693.

14. Kredentser MS, Martens PJ, Chochinov HM, Prior HJ. Cause and rate of death in people with schizophrenia across the lifespan: a population-based study in Manitoba, Canada. The Journal of clinical psychiatry. 2014;75(2):154–61. Epub 2014/03/08. pmid:24602251.

15. Maayan L, Correll CU. Management of antipsychotic-related weight gain. Expert review of neurotherapeutics. 2010;10(7):1175–200. Epub 2010/07/01. pmid:20586697; PubMed Central PMCID: PMC3501406.

16. Raben AT, Marshe VS, Chintoh A, Gorbovskaya I, Muller DJ, Hahn MK. The Complex Relationship between Antipsychotic-Induced Weight Gain and Therapeutic Benefits: A Systematic Review and Implications for Treatment. Frontiers in neuroscience. 2017;11:741. Epub 2018/02/07. pmid:29403343; PubMed Central PMCID: PMC5786866.

17. Panariello F, De Luca V, de Bartolomeis A. Weight gain, schizophrenia and antipsychotics: new findings from animal model and pharmacogenomic studies. Schizophrenia research and treatment. 2011;2011:459284. Epub 2011/01/01. pmid:22988505; PubMed Central PMCID: PMC3440684.

18. Kanter R, Caballero B. Global gender disparities in obesity: a review. Adv Nutr. 2012;3(4):491–8. Epub 2012/07/17. pmid:22797984; PubMed Central PMCID: PMC3649717.

19. Gebhardt S, Theisen FM, Haberhausen M, Heinzel-Gutenbrunner M, Wehmeier PM, Krieg JC, et al. Body weight gain induced by atypical antipsychotics: an extension of the monozygotic twin and sib pair study. Journal of clinical pharmacy and therapeutics. 2010;35(2):207–11. Epub 2010/05/12. pmid:20456740.

20. Anders S, Huber W. Differential expression analysis for sequence count data. Genome biology. 2010;11(10):R106. Epub 2010/10/29. pmid:20979621; PubMed Central PMCID: PMC3218662.

21. Trapnell C, Pachter L, Salzberg SL. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics. 2009;25(9):1105–11. Epub 2009/03/18. pmid:19289445; PubMed Central PMCID: PMC2672628.

22. Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome biology. 2009;10(3):R25. Epub 2009/03/06. pmid:19261174; PubMed Central PMCID: PMC2690996.

23. Quinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics. 2010;26(6):841–2. Epub 2010/01/30. pmid:20110278; PubMed Central PMCID: PMC2832824.

24. Mitchell JA, Aronson AR, Mork JG, Folk LC, Humphrey SM, Ward JM. Gene indexing: characterization and analysis of NLM's GeneRIFs. AMIA Annual Symposium proceedings / AMIA Symposium AMIA Symposium. 2003:460–4. Epub 2004/01/20. pmid:14728215; PubMed Central PMCID: PMC1480312.

25. Croft D, O'Kelly G, Wu G, Haw R, Gillespie M, Matthews L, et al. Reactome: a database of reactions, pathways and biological processes. Nucleic acids research. 2011;39(Database issue):D691–7. Epub 2010/11/12. pmid:21067998; PubMed Central PMCID: PMC3013646.

26. Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic acids research. 2000;28(1):27–30. Epub 1999/12/11. pmid:10592173; PubMed Central PMCID: PMC102409.

27. Slenter DN, Kutmon M, Hanspers K, Riutta A, Windsor J, Nunes N, et al. WikiPathways: a multifaceted pathway database bridging metabolomics to other omics research. Nucleic acids research. 2018;46(D1):D661–D7. Epub 2017/11/15. pmid:29136241; PubMed Central PMCID: PMC5753270.

28. Schaefer CF, Anthony K, Krupa S, Buchoff J, Day M, Hannay T, et al. PID: the Pathway Interaction Database. Nucleic acids research. 2009;37(Database issue):D674–9. Epub 2008/10/04. pmid:18832364; PubMed Central PMCID: PMC2686461.

29. Bak M, Fransen A, Janssen J, van Os J, Drukker M. Almost all antipsychotics result in weight gain: a meta-analysis. PloS one. 2014;9(4):e94112. Epub 2014/04/26. pmid:24763306; PubMed Central PMCID: PMC3998960.

30. Correll CU, Manu P, Olshanskiy V, Napolitano B, Kane JM, Malhotra AK. Cardiometabolic risk of second-generation antipsychotic medications during first-time use in children and adolescents. Jama. 2009;302(16):1765–73. Epub 2009/10/29. pmid:19861668; PubMed Central PMCID: PMC3055794.

31. Safer DJ. A comparison of risperidone-induced weight gain across the age span. Journal of clinical psychopharmacology. 2004;24(4):429–36. Epub 2004/07/03. pmid:15232335.

32. Eranti SV, MacCabe JH, Bundy H, Murray RM. Gender difference in age at onset of schizophrenia: a meta-analysis. Psychological medicine. 2013;43(1):155–67. Epub 2012/05/09. pmid:22564907.

33. Lee HS, Barraza-Villarreal A, Biessy C, Duarte-Salles T, Sly PD, Ramakrishnan U, et al. Dietary supplementation with polyunsaturated fatty acid during pregnancy modulates DNA methylation at IGF2/H19 imprinted genes and growth of infants. Physiological genomics. 2014;46(23):851–7. Epub 2014/10/09. pmid:25293351; PubMed Central PMCID: PMC4254937.

34. Nomura I, Kato J, Tokashiki M, Kitamura K. Increased plasma levels of the mature and intermediate forms of adrenomedullin in obesity. Regulatory peptides. 2009;158(1–3):127–31. Epub 2009/08/27. pmid:19706311.

35. Friedrichsen M, Poulsen P, Wojtaszewski J, Hansen PR, Vaag A, Rasmussen HB. Carboxylesterase 1 gene duplication and mRNA expression in adipose tissue are linked to obesity and metabolic function. PloS one. 2013;8(2):e56861. Epub 2013/03/08. pmid:23468884; PubMed Central PMCID: PMC3585247.

36. Mortensen OH, Nielsen AR, Erikstrup C, Plomgaard P, Fischer CP, Krogh-Madsen R, et al. Calprotectin—a novel marker of obesity. PloS one. 2009;4(10):e7419. Epub 2009/10/14. pmid:19823685; PubMed Central PMCID: PMC2758581.

37. Phillips CM, Goumidi L, Bertrais S, Field MR, Cupples LA, Ordovas JM, et al. Gene-nutrient interactions with dietary fat modulate the association between genetic variation of the ACSL1 gene and metabolic syndrome. Journal of lipid research. 2010;51(7):1793–800. Epub 2010/02/24. pmid:20176858; PubMed Central PMCID: PMC2882737.

38. Gupta A, Rezvani R, Lapointe M, Poursharifi P, Marceau P, Tiwari S, et al. Downregulation of complement C3 and C3aR expression in subcutaneous adipose tissue in obese women. PloS one. 2014;9(4):e95478. Epub 2014/04/20. pmid:24743347; PubMed Central PMCID: PMC3990631.

39. Ha E, Kim MJ, Choi BK, Rho JJ, Oh DJ, Rho TH, et al. Positive association of obesity with single nucleotide polymorphisms of syndecan 3 in the Korean population. The Journal of clinical endocrinology and metabolism. 2006;91(12):5095–9. Epub 2006/10/05. pmid:17018662.

40. Wang C, Ha X, Li W, Xu P, Gu Y, Wang T, et al. Correlation of A2bAR and KLF4/KLF15 with Obesity-Dyslipidemia Induced Inflammation in Uygur Population. Mediators of inflammation. 2016;2016:7015620. Epub 2016/05/21. pmid:27199507; PubMed Central PMCID: PMC4856914.

41. Turcot V, Bouchard L, Faucher G, Tchernof A, Deshaies Y, Perusse L, et al. A polymorphism of the interferon-gamma-inducible protein 30 gene is associated with hyperglycemia in severely obese individuals. Human genetics. 2012;131(1):57–66. Epub 2011/06/28. pmid:21701784.

42. Shi J, Zhang Y, Gu W, Cui B, Xu M, Yan Q, et al. Serum liver fatty acid binding protein levels correlate positively with obesity and insulin resistance in Chinese young adults. PloS one. 2012;7(11):e48777. Epub 2012/11/13. pmid:23144966; PubMed Central PMCID: PMC3492433.

43. Vehof J, Risselada AJ, Al Hadithy AF, Burger H, Snieder H, Wilffert B, et al. Association of genetic variants of the histamine H1 and muscarinic M3 receptors with BMI and HbA1c values in patients on antipsychotic medication. Psychopharmacology. 2011;216(2):257–65. Epub 2011/02/22. pmid:21336576; PubMed Central PMCID: PMC3121946.

44. Stefansson H, Sigurdsson E, Steinthorsdottir V, Bjornsdottir S, Sigmundsson T, Ghosh S, et al. Neuregulin 1 and susceptibility to schizophrenia. American journal of human genetics. 2002;71(4):877–92. Epub 2002/07/30. pmid:12145742; PubMed Central PMCID: PMC378543.

45. Lee W. The CEACAM1 expression is decreased in the liver of severely obese patients with or without diabetes. Diagnostic pathology. 2011;6:40. Epub 2011/05/17. pmid:21569294; PubMed Central PMCID: PMC3104481.

46. Wang JB, Zhang YJ, Guan J, Zhou L, Sheng Y, Zhang Y, et al. Enhanced syndecan-1 expression on neutrophils in patients with type 2 diabetes mellitus. Acta diabetologica. 2012;49(1):41–6. Epub 2011/02/18. pmid:21327984.

47. Monroy-Munoz IE, Angeles-Martinez J, Posadas-Sanchez R, Villarreal-Molina T, Alvarez-Leon E, Flores-Dominguez C, et al. PLA2G2A polymorphisms are associated with metabolic syndrome and type 2 diabetes mellitus. Results from the genetics of atherosclerotic disease Mexican study. Immunobiology. 2017;222(10):967–72. Epub 2016/09/10. pmid:27608594.

48. Kume T, Acar S, Tuhan H, Catli G, Anik A, Gursoy Calan O, et al. The Relationship between Serum Zonulin Level and Clinical and Laboratory Parameters of Childhood Obesity. Journal of clinical research in pediatric endocrinology. 2017;9(1):31–8. Epub 2016/12/23. pmid:28008865; PubMed Central PMCID: PMC5363162.

49. Kim JY, Campbell LE, Shaibi GQ, Coletta DK. Gene expression profiling and association of circulating lactoferrin level with obesity-related phenotypes in Latino youth. Pediatric obesity. 2015;10(5):338–44. Epub 2014/11/15. pmid:25394788.

50. Bradfield JP, Taal HR, Timpson NJ, Scherag A, Lecoeur C, Warrington NM, et al. A genome-wide association meta-analysis identifies new childhood obesity loci. Nature genetics. 2012;44(5):526–31. Epub 2012/04/10. pmid:22484627; PubMed Central PMCID: PMC3370100.

51. Corripio R, Gonzalez-Clemente JM, Perez-Sanchez J, Naf S, Gallart L, Nosas R, et al. Weight loss in prepubertal obese children is associated with a decrease in adipocyte fatty-acid-binding protein without changes in lipocalin-2: a 2-year longitudinal study. European journal of endocrinology. 2010;163(6):887–93. Epub 2010/09/28. pmid:20870707.

52. Madrigal-Ruiz PM, Navarro-Hernandez RE, Ruiz-Quezada SL, Corona-Meraz FI, Vazquez-Del Mercado M, Gomez-Banuelos E, et al. Low CD36 and LOX-1 Levels and CD36 Gene Subexpression Are Associated with Metabolic Dysregulation in Older Individuals with Abdominal Obesity. Journal of diabetes research. 2016;2016:5678946. Epub 2016/08/16. pmid:27525284; PubMed Central PMCID: PMC4976145.

53. Lauhio A, Farkkila E, Pietilainen KH, Astrom P, Winkelmann A, Tervahartiala T, et al. Association of MMP-8 with obesity, smoking and insulin resistance. European journal of clinical investigation. 2016;46(9):757–65. Epub 2016/06/15. pmid:27296149.

54. McAinch AJ, Cornall LM, Watts R, Hryciw DH, O'Brien PE, Cameron-Smith D. Increased pyruvate dehydrogenase kinase expression in cultured myotubes from obese and diabetic individuals. European journal of nutrition. 2015;54(7):1033–43. Epub 2014/10/15. pmid:25311062.

55. Elmose C, Sverrild A, van der Sluis S, Kyvik KO, Backer V, Thomsen SF. Genetic factors explain half of all variance in serum eosinophil cationic protein. Clinical and experimental allergy: journal of the British Society for Allergy and Clinical Immunology. 2014;44(12):1525–30. Epub 2014/10/30. pmid:25354326.

56. Ejchel TF, Araujo LM, Ramos LR, Cendoroglo MS, de Arruda Cardoso Smith M. Association of the apolipoprotein A-IV: 360 Gln/His polymorphism with cerebrovascular disease, obesity, and depression in a Brazilian elderly population. American journal of medical genetics Part B, Neuropsychiatric genetics: the official publication of the International Society of Psychiatric Genetics. 2005;135B(1):65–8. Epub 2005/04/05. pmid:15806598.

57. Kabaroglu C, Onur E, Barutcuoglu B, Ozhan B, Erdinc S, Var A, et al. Inflammatory marker levels in obese adolescents with glucose intolerance: increased chitotriosidase activity. Clinical biochemistry. 2012;45(4–5):281–4. Epub 2011/12/31. pmid:22206738.

58. Haas E, Bhattacharya I, Brailoiu E, Damjanovic M, Brailoiu GC, Gao X, et al. Regulatory role of G protein-coupled estrogen receptor for vascular function and obesity. Circulation research. 2009;104(3):288–91. Epub 2009/01/31. pmid:19179659; PubMed Central PMCID: PMC2782532.

59. Aslibekyan S, Do AN, Xu H, Li S, Irvin MR, Zhi D, et al. CPT1A methylation is associated with plasma adiponectin. Nutrition, metabolism, and cardiovascular diseases: NMCD. 2017;27(3):225–33. Epub 2017/02/01. pmid:28139377; PubMed Central PMCID: PMC5330786.

60. Patel PS, Buras ED, Balasubramanyam A. The role of the immune system in obesity and insulin resistance. Journal of obesity. 2013;2013:616193. Epub 2013/04/12. pmid:23577240; PubMed Central PMCID: PMC3618935.

61. Hotamisligil GS. Inflammation and metabolic disorders. Nature. 2006;444(7121):860–7. Epub 2006/12/15. pmid:17167474.

62. Dandona P, Aljada A, Bandyopadhyay A. Inflammation: the link between insulin resistance, obesity and diabetes. Trends in immunology. 2004;25(1):4–7. Epub 2003/12/31. pmid:14698276.